Democritus University of Thrace (DUTH)
Ανάπτυξη συστήματος υποστήριξης αποφάσεων για τον χρονοπρογραμματισμό έργων με χρήση μεταευρετικών και πολυσκοπικών αλγόριθμων
Abstract
dc:descriptionThe aim of this doctoral thesis is to develop a Decision Support System (DSS) for addressing project scheduling problems. In DSS, both new scheduling models and new computational methods are included. The algorithms used into the DSS were chosen after extensive computational analysis of their performance. The doctoral thesis is divided into nine chapters. In Chapter 1 the concept of the project is defined and the important role of project management in project successful realization is explained. The concept of the DSS is defined, its structure is described and project management oriented DSS are reported. Finally, the DSS structure developed in this thesis is outlined. In Chapter 2 concepts of single-objective and multi-objective optimization are introduced and basic concepts relating to their complexity are mentioned, and the main features of heuristics and metaheuristics algorithms are described. Specific reference is made to Pareto metaheuristics algorithms for addressing multi-objective problems, especially to the three Pareto metaheuristics algorithms PAES, PESA2 and NSGA2, which are used in most of the problems of the thesis. Chapter 3 describes in detail the computational analysis applied for the evaluation of algorithms. The factors taken into account in the computational analysis are specified, the construction of the reference dataset in assessing algorithms is described, the criteria for evaluating the algorithms are analyzed, and the statistical techniques used to extract the conclusions. Chapter 4 focuses on the problem of the time/cost trade-off in a project whose activities can be performed with discrete time/cost combinations. The general form of the problem is resolved by the three Pareto metaheuristics algorithms, while algorithms for the three specific cases of the problem arc proposed -minimizing the direct cost of the project under a given duration of the project, minimizing project duration under a given budget, and minimizing the overall cost of the project. Chapter 5 proposes a model to optimize the quality of project scheduling with discrete time/cost combinations. The problem is formulated as a multi-objective optimization problem with three objective functions, the goal being the time/cost/robustness trade-off of the schedule, and is solved by the three Pareto algorithms. Chapter 6 presents a new model for resource leveling. The problem is treated as multi-objective -minimizing simultaneously resource leveling and the duration of the project, and is resolved by the three Pareto metaheuristics. Chapter 7 is devoted to the construction of the activity-on-arc representation of a project that is given the activity-on-node representation of a project, construct its activity-on-arc representation. The solution procedure is based on an algorithm known from literature for solving the NP-hard dummy-arc problem, i.e. an activity-on-arc representation of a project that has the minimum number of dummy arcs given that it has the minimum number of nodes. For the solution of the resulting set covering problem two variants of a Greedy Randomized Adaptive Search Procedure (GRASP) are proposed. Chapter 8 describes and compares three approaches for dealing with stochastic networks: the Monte Carlo simulation, the PERT method and the modelization of project scheduling via fuzzy sets. Chapter 9 contains the conclusions and issues for further research. Appendix describes the Decision Support System, which was coded in Visual Basic 6.0 and operates under Microsoft Project.
Degree
thesis:*- Grantor dc:publisher
- Democritus University of Thrace (DUTH)
- Year dc:date
- 2008
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Kotsikas, L.
- Κώτσικας, Λάζαρος
Subjects
dc:subject × 26- Εξομάλυνση πόρων
- Ευρωστία χρονοδιαγράμματος
- Μεταευρετικοί αλγόριθμοι
- Πολυσκοπική βελτιστοποίηση
- Στοχαστικά δίκτυα έργων
- Συστήματα υποστήριξης αποφάσεων
- Χρονοπρογραμματισμός έργων
- Pareto βελτιστοποίηση
- Resource leveling
- Project robustness
- Metaheuristics
- Multi-objective optimization
- Stochastic networks
- Decision support systems
- Project scheduling
- Pareto optimization
- Επιστήμη Πολιτικού Μηχανικού
- Επιστήμες Μηχανικού και Τεχνολογία
- Κατασκευή και Οικοδομική τεχνολογία
- Άλλες Επιστήμες Μηχανικού και Τεχνολογίες
- Μηχανική και Τεχνολογίες, άλλοι τομείς
- Civil Engineering
- Engineering and Technology
- Construction and Building technology
- Other Engineering and Technologies
- Engineering and Technologies, miscellaneous
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/20983